substitution.
Step 4. She revises her hypothesis and systematically replaces the bat wing ingredient and the poppy seed ingredient.
Step 5. She observes that the townspeople no longer report that they feel better after visiting the witch when the poppy seeds have been replaced.
Step 6. She concludes that poppy seeds have a medicinal quality that promotes feelings of well-being.
Theories, Concepts, and Hypotheses
The objectives of science can be seen in theories. We use theories to describe what is known in an area, present an explanation of those findings, and make predictions for further research. A theory is a formal statement of how concepts are related. Concepts are the general category of ideas that are represented by our variables. Theories may be very general and account for many phenomena, such as Skinner’s behavioral theory, with applications to all of human behavior, or more specific and limited in scope, such as Epling and Pierce’s theory of activity-based anorexia.
If we were all-knowing, we would not need theories. We would know how the universe worked, and research would be predictable and boring. Fortunately for those of us who enjoy research, we do not have all the answers, so we construct theories of how we think the world works. The main advantage of a theory is that it provides an explanation of how concepts are related. So rather than having to remember a whole library of specific research findings, we need only to remember and apply the theory. The theory will describe how general concepts are related.
Theories are an integral part of the research process. In addition to explaining what we already know, we use theories to make new predictions that can be empirically tested. By using specific instances of the general concepts, we can derive new testable hypotheses. A hypothesis is a prediction of how concepts are related that is often deduced from a theory. We then conduct our research to test the hypothesis. If the hypothesis is supported by the research, the theory is strengthened. On the other hand, if the hypothesis is not supported, the theory may have to be altered. Theories do not live forever. They start out wonderfully, explaining and organizing a whole collection of observations. Over their life, they gain support from some research, they may make surprising new predictions, and they may fail to explain some research findings. When enough research is compiled that does not fit the theory, a new theory will be proposed.
So what is the nature of empirical research? Where do scientists get their ideas, and how do they go about meeting their goals to describe, explain, predict, and control phenomena? Let’s look at five common reasons researchers might have for conducting research.
Why We Do Research
To Evaluate a Theory
In psychology, theories abound. Theories, if they are good theories, generate testable hypotheses. Good theories allow us to test the hypotheses derived from them. Bad theories often do not. In fact, one criterion of a good theory is whether testable hypotheses can be postulated. A great deal of research in psychology is conducted to evaluate current theories about human behavior. In a classic article, Darley and Latane (1968) offered a theory about why the many people who could hear, from their apartments, a young woman being attacked outside did nothing to help her. They postulated that the responsibility to be good citizens was diffused among the many people, and as a result, no one person felt compelled to help. One hypothesis that can be derived from this theory (called the bystander effect) is that the greater the number of people present, the less likely any one person is to help someone in distress. And, indeed, this hypothesis has been confirmed in numerous experiments.
Let’s look at another example. Developmental psychologists call the emotional bond between children and their primary caregivers attachment. But why does this happen? Behaviorists proposed that attachment develops because the primary caregiver, usually the mother, is associated with food, a strong positive reinforcer to hungry babies. Harlow and Harlow (1966) tested this theory in a classic set of studies. Their results did not support the behaviorists’ claim. Contact comfort, not feeding, was shown to be the source of attachment, at least in rhesus monkeys.
Theories provide a wealth of ideas for research topics.
To Satisfy Our Curiosity
Science often develops because scientists are very curious people. We have heard it said that you could get rid of warts by visiting a graveyard around midnight. We will not provide all the details of this activity, but chasing away evil spirits, and presumably your wart, by tossing a cat is involved (see The Adventures of Tom Sawyer). Spanos, Williams, and Gwynn (1990) were curious about the idea that you can rid yourself of warts by nonmedical means; they decided to investigate whether you could get rid of warts through hypnosis. They conducted a controlled experiment in the laboratory and found support for the hypothesis. They published their results in a scientific journal. But they were more than a little embarrassed when their study made the cover of The National Enquirer—not exactly something you want to brag about at your university!
To Demonstrate a New Technique
As we learn new ways to do things, it is important to determine if those new ways are better than the old ways. Professors are always interested in better ways of teaching. Textbooks these days come with all sorts of fancy supplementary materials. But do they improve learning? At our school, we conducted an experiment to try to answer that question. Different groups of students taking introductory psychology received instruction with various technological accompaniments. Some received traditional lecture instruction. Some received computer-assisted instruction. We measured several variables, including performance and more psychological variables. Are you wondering what we found? Well, very simply, the students with added technology did not learn more, but they had more fun!
To Demonstrate a Behavioral Phenomenon
After observing behavior that tends to recur under certain circumstances, we need to demonstrate it under precise conditions before the phenomenon can be confidently added to the body of knowledge about a discipline. The idea that organisms do things because they receive rewards was known for a long time before Skinner demonstrated this in his laboratory. Indeed, Skinner’s career was spent demonstrating the behavioral phenomena of operant conditioning.
Perhaps you have noticed that when you are out walking your dog, people seem a lot friendlier than when you are out walking alone. You could design a simple experiment to see if your perception that people are friendlier is a demonstrable phenomenon. You could take the same walk at the same time each day for several weeks, sometimes with your dog and sometimes without. You might collect data on how many people engage you in conversation and how long they talk to you, for example. If you find that when your dog is with you, more people initiate conversation more often, you have evidence of a behavioral phenomenon.
To Investigate the Conditions Influencing Behavioral Phenomena
Darley and Latane’s (1968) bystander effect has been the focus of numerous experiments. Researchers have studied not only the influence of the number of bystanders on helping behavior but also many other factors, such as the apparent degree of need of the victim and the bystanders’ ability to help. Skinner and others have investigated how the size, frequency, and quality of a reward (reinforcer) affect behavior, as well as many other conditions affecting operant behavior.
We have discussed some of the reasons why researchers do what they do. Now let’s discuss the various ways they go about doing what they do.
Approaches to Research
Over many years of schooling, students are trained to be convergent thinkers, to converge on the one correct answer. But research requires divergent thinking. It is a creative endeavor with many approaches. Here we summarize the diversity of research by organizing various approaches on a number of typical continua you have probably come across in your undergraduate